It doesn't matter whether the computer makes the best move or not, as long as your move is the best and leads to an advantage after any reply.
If white's first move means black's best is to sacrifice his queen, then the problem works.
If the tactic requires a bad move for it to work, then the problem is broken.
@OP: Nevertheless, if you think you have indeed found a flawed TT problem, you can report it to Help & Support. Go to this link:
http://support.chess.com/Tickets/Submit
and click the radio button for Site Feedback & Suggestions.
Thanks, whirlwind.
I know it doesn't matter if the computer makes the strongest move, Scottrf, but the analysis is showing people that the weakest potential move is the strongest on that one. The strongest being painted as the weakest. The most common reply to questions is look at the analysis. If the analysis is flawed on a problem, then I'm not sure how that really helps someone learn a strong chess game. Especially when they get trained to make moves that lead to opponent advantage levels that indicate it's approaching an avoidable or at least delayable extended checkmate.
Ah, so the analysis is wrong, not the tactic.
Pretty much! Since the analysis is used to make the tactic moves, that makes the tactic wrong too.
It doesn't make the tactic wrong though, because the tactic attempts to provide the most testing response, not the best.
It would be a worse tactic if they used the queen sacrifice because everyone will know to take the queen. By moving the king away you have to prove the point of the initial move.
@OP: You're welcome! Although I must admit that with advantages of 12 versus 21 points, we really needn't be concerned with whether moves are "painted" as stronger or weaker, because the advantage is insurmountable in either case.
@Scottrf: I agree, but I was merely letting be known how to report any flawed TT problem, should one be found.
I just ran into an unexpected result in Problem #0036566 when I ran my own analysis on the board after reading questions in the comments. According to the problem flow and analysis, black made the strongest move out of three that avoid checkmate. However, it's unclear which move is strongest for black until a deeper search, at which point it becomes the weakest at an advantage for white of over +21. The originally considered weakest move that sacrifices black's queen becomes the strongest by at least 12 score.